Algorithms Processing

The simulation parameters for the processing of algorithms are given in the table below. The data that travels from transmitter to receiver follows the flow chart given below. The Random data was generated by setting the number of subcarriers, modulation scheme and OFDM symbols. The modulation schemes according to Table 3 were used and CP of length 8 was also used to remove the ISI from data. Training bits were concatenated with IFFT data of size 2048 to perform equalization at the receiver end.

Table 3. Simulation Parameters

Parameters Values

IFFT Size 2048

Number of Subcarriers 1200

Modulation Schemes QPSK,16-QAM, 32-QAM, 64-QAM

Signal to Noise Ratio -30:2:20 dB

Number of Transmit Antennas 64

Number of Receive Antennas 1,2,4,6

Cyclic-Prefix 8

A single stream of data is generated and transmitted from each antenna. Singular Value Decomposition (SVD) method is used to optimize the MIMO channel capacity.

Figure 22. Working of Beamforming Algorithms in CP-OFDM Process [39][40] The Transmitted signal passed through the MIMO channel where noise was added. Maximum Ratio Combining (MRC) combined the individual streams for each receiver antenna one by one according to individual SNR. On the receiver side, CP was removed and FFT of the data performed. The received data was equalized with Zero-Forcing equalization by inversing the frequency response of the channel. The bit stream before the modulation and the after demodulation are compared to check the BER as shown in the Figure 22. The simulation parameters as shown in Table 3 were varied in different use cases to evaluate the BER.

Figure 23. The Figure shows the BER vs SNR plot for SISO Channel with Modula-

tion scheme QAM=16, Subcarriers=1200, CP=8 and OFDM Symbols=15, numtx=1

The beamforming algorithms are used in the SISO channel to measure the BER perfor- mance with respect to SNR. From Figure 23 when the single antenna system is config- ured in the CP-OFDM system with MRC and ZF algorithms on the transmitter and re- ceiver side respectively, the BER value is becoming better and better. As SNR values are increased in the signal, the BER curve continuously decreasing. For instance, at SNR=-20 dB the value of BER is, BER=0.49 and at SNR=0 dB, BER=0.3385 and simi- larly, it decreases in the same fashion as SNR increases, so at SNR=28dB, it is BER=0.0015. So, before simulation and after, it is clear that the performance with a single antenna configuration is improving as SNR value is increasing.

Figure 24. The Figure shows the BER vs SNR plot for MISO channel with different

Modulation schemes QAM=4, QAM=16, QAM=32, QAM=64, Subcarriers=1200, cp=8, OFDM Symbols=15, numtx=64 and numrx=1

Figure 24 illustrates the BER performance comparison of different modulation schemes keeping simulation parameters such as antennas, OFDM symbols, subcarriers and CP length fixed. Different modulation schemes were implemented to evaluate the beamform- ing algorithms. It is shown is in the above figure that as modulation schemes are going higher from QAM-4 to QAM-64, performance start to decrease. The QAM=64 gave very high BER values when comparing with other modulation schemes. At SNR=-20 dB, BER=0.49, all modulation schemes gave the same performance. But as SNR values increased, BER value decreased for all schemes. At SNR=0dB, BER=0.2249,0.3278, 0.3690,0.3917 for QAM=4,16,32 and QAM=64 respectively. When we further increased the SNR values then the performance difference between all schemes became huge. At SNR=16dB, BER=0,0.0078,0.0369,0.0723 for QAM=4,16,32 and 64 respectively. After a comparison of all modulation schemes, it is concluded that QAM=64 at high noise gave worst performance. On the other hand, QAM-4 showed good performance.

Figure 25. The Figure shows the BER vs SNR plot for MIMO channel with different

Receive antennas, numrx=1,2,4,6, QAM=16, Subcarriers=1200, cp=8, OFDM Sym- bols=15 and numtx=64

Figure 25 illustrates the BER performance comparison of different receive antennas con- sidering fixed simulation parameters such as modulation scheme, transmit antennas, OFDM symbols, subcarriers, and CP length. A use case was made where different re- ceive antennas were implemented to evaluate the beamforming algorithms in the CP- OFDM system. At SNR=-20 dB, BER=0.4977, all receive antennas got the same power and BER was also the same. As SNR values increased BER value decreased for all receive antennas. At SNR=0dB, BER=0.3415,0.2633, 0.1803,0.1326 for numrx=1,2,4 and numrx=6 respectively. When we further increased the SNR values then performance difference between all receive antennas became large. At SNR=20 dB, BER= 2.7e- 05,0,0,0 for numrx=1,2,4 and 6 respectively. After a comparison of all receive schemes, it is concluded that from all receive antennas, numrx=6 gave much better performance in CP-OFDM system as compared to other antennas.

Figure 26. The Figure shows the BER vs SNR plot for MIMO channel with different

Transmit antennas, numtx=1,2,4,6, QAM=16, Subcarriers=1200, cp=8, OFDM Sym- bols=15 and numrx=64

Figure 26 illustrates the BER performance comparison of a different transmit antennas considering fixed simulation parameters such as modulation scheme, receive antennas, OFDM symbols, subcarriers, and CP length. Different transmit antennas were imple- mented to evaluate the beamforming algorithms. It is seen from Figure 26 that all transmit antennas at the same SNR values gave the same BER values. So, we can say that by increasing the SNR values and providing equal transmit power, all transmit antennas had same BER in the CP-OFDM environment.